Our Federated Learning AI models are based on data from multiple affiliated hospitals, instead of a single department of a single hospital.
Our AI tool is developed based on a combination of Federated Learning (FL) and secure Multi Party Computation (MPC) that ensures that the data remains at the source as much as possible, and that intermediate models are aggregated in a secure manner. Software has been developed that makes it possible to make data available for data analysis and modeling in a secure and privacy-protecting manner. The algorithm in the AI tool queries the data using FL, which makes it possible for the AI tool to learn from the data, without the data actually having to be shared. This makes it possible to use data from multiple sources (for example, multiple hospitals) to train the intended AI models. And all this in a clear and very user-friendly way using the generative LLMs